All opinions expressed herein are those of the authors and should not be reproduced, quoted in publications, or
used as a reference without the author’s consent.
Oak Ridge National Laboratory is managed by UT-Battelle, LLC, for the U.S. Department of Energy.
Published by Fusion Energy Division, Oak Ridge National Laboratory
Building 5700 P.O. Box 2008 Oak Ridge, TN 37831-6169, USA
Editor: James A. Rome Issue 133 August 2011
E-Mail: jamesrome@gmail.com Phone (865) 482-5643
On the Web at http://www.ornl.gov/sci/fed/stelnews
Design and installation of a
test helical divertor with a
baffle structure in LHD
The helical divertor in the Large Helical Device (LHD) at
the National Institute for Fusion Research (NIFS) in Toki,
Japan was partially modified before the 2010 experimental
campaign to demonstrate control of particles by using a
baffle structure. This baffle structure consists of watercooled
divertor plates combined with baffle plates and a
dome inside the private region of the divertor. The recycled
neutral gas was successfully compressed in the modified
divertor during discharges, and more than ten times
higher pressure was observed there than in the unmodified
divertor.
The control of neutral particles using a divertor is a crucial
issue for a fusion reactor such as the force-free helical
reactor (FFHR) [1]. Hydrogen isotopes, helium ash, and
impurities must be pumped to sustain the burning fusion
plasma. Closed divertor experiments have also been conducted
in the Compact Helical System (CHS) [2], Wendelstein-
7AS [3], and LHD [4] using a magnetic island. In
LHD, a local island divertor (LID) experiment was conducted,
and effective particle control was demonstrated
[5]. Furthermore, the superdense core (SDC) plasma operational
regime caused by the formation of an internal diffusion
barrier (IDB) was obtained during high pumping
operation using the LID [6]. However, the wetted area on
the divertor head of the LID was so small that it could not
be utilized during long-pulse operation in a high-performance
plasma. The wetted area in the helical divertor,
which is an inherent part of the heliotron magnetic configuration,
is larger than that in the LID. However, the neutral
pressure in the helical divertor only reaches about 0.01 Pa
even during a high-density discharge in which the lineaveraged
density is higher than 1020 m3, and it is
believed that the neutral pressure must be ten times higher
for effective particle control using divertor pumping [7].
The relatively low neutral pressure in the helical divertor
was considered to be caused by the three-dimensional
(3D) plasma distribution in the divertor and the large volume
of the vacuum vessel. Therefore it is necessary to
install a baffle structure to compress neutral particles in
the divertor region.
The baffle structure for the helical divertor was designed
using a fully 3D neutral transport simulation [8] with the
EIRENE code [9]. The code was applied to study the
behavior of neutrals in the existing divertor configuration
before the start of the design, and the results of the calculation
were validated by comparing them to experimental
results obtained by spectroscopic measurements [10].
The development of the plasma-facing components for the
divertor was conducted using the electron beam irradiation
device, ACT (Active Cooling Test stand) at NIFS, on the
basis of a mechanically jointed water-cooled divertor plate
used in the existing divertor [7, 11]. The divertor plate’s
temperature is considered to be highest during a steadystate
discharge using 3 MW heating power, which is an
experimental goal of LHD. In such discharges, the largest
heat flux to divertor plates is estimated to be 1.5 MW/m2,
and the divertor plates were developed to tolerate this heat
load. The design of the divertor and the development of
the plasma-facing components were completed in 2008
[7], and two of ten torus inboard side divertors were modi-
In this issue . . .
Design and installation of a test helical divertor
with a baffle structure in LHD
The Large Helical Divertor experiment has designed
and installed a test helical divertor in LHD and operated
it in the 2010 experimental campaign. In the new
divertor, the plates face into the private region, and a
protective dome is used to plug up the throat region.
The divertor was designed to provide control of the
neutral pressure during high-power neutral beam
injection. During injection, the neutral pressure was an
order of magnitude higher than in the unmodified
divertor. After the 2011 campaign, this divertor design
will be installed in an additional six toroidal sections of
LHD. ....................................................................... 1
Stellarator News -2- August 2011
fied by installing the baffle structure before the 2010
experimental campaign to demonstrate neutral particle
compression in the helical divertor [12].
Figure 1 shows a schematic view of both the existing and
the modified divertor. In the existing divertor, the plasmafacing
surfaces of the divertor plates face the main plasma.
In the modified divertor, these surfaces face the private
region which also contains the “dome” structure. Due to
the changes of the divertor plates’ angles and the installation
of the dome, it was expected from the neutral transport
calculations that the neutral pressure under the dome
should be at least ten times higher than that in the private
region in the existing divertor [7, 8], as shown in Fig. 2.
Figure 3 shows the modified divertor. One of the modified
divertors provides armor to protect the vacuum vessel
against the neutral beam (NB) used for plasma heating.
The divertor plates and dome are made from isotropic
graphite, and they are water cooled by mechanical jointing
to water cooling pipes made of SUS316 [7, 11]. Carbonfiber
composite is applied to the divertor plates and dome
components in the NB deposition area.
Figure 4 shows a CCD camera view of the modified divertor
during a discharge. In the photo, H emission is intense
in the region between divertor plates and dome, suggesting
that neutral particles interact intensely with plasma in the
region.
Fig. 1. Schematic views of an unmodified (top) and a modified
divertor (bottom) in the torus inboard side. The righthand
figures are the cross-sectional views of the divertors
in the equatorial plane. [12]
Fig. 2. Comparison of the neutral pressure in the divertor
region of the unmodified and modified helical divertors.
Fig. 3. Photographs of the modified divertor.
Stellarator News -3- August 2011
To measure the neutral pressure in the modified divertor,
three fast-ion gauges were installed under the dome. To
compare the pressure in the modified and unmodified
divertors, another fast ion gauge was installed in the torus
inboard-side private region of an unmodified divertor.
These fast ion gauges were calibrated using a cold cathode
gauge and a capacitance gauge from 5 ×103 to 2 Pa.
Density ramp-up discharges were conducted in a standard
magnetic configuration in which the magnetic axis radius
is 3.6 m to compare the neutral pressures in the modified
and the unmodified divertors. Figure 5 shows the time
evolutions of plasma stored energy, line-averaged electron
density (ne,bar) in the center chord, and the neutral pressure
near the equatorial plane in these divertors during a density
ramp-up discharge. Hydrogen gas puffing was conducted
during the discharge, and ne,bar increased to 7
×1019 m3. The neutral pressures in both divertors
increased with increasing ne,bar, and the pressure in the
modified divertor was more than 10 times higher than that
in the unmodified divertor throughout the discharge. After
the termination of the NB injection, the neutral pressure in
the modified divertor decreased and become the same as
that in the unmodified divertor. This result proves that
neutral particle compression in the modified divertor
works well.
No negative effects on the plasma operation caused by the
installed baffle structure have thus far been observed. The
initial result promises that an upgraded divertor system
with pumps under the dome will provide efficient active
particle control in LHD.
The 2011 experimental campaign began on 28 July without
any additional modifications in the divertor region.
After the campaign, starting in November, we are going to
modify the divertor in another six toroidal sections and
install in-vessel cryopumps under their domes.
This work was supported by a Grant-in-Aid for Scientific
Research (A) (20246135), and the NIFS budget under contracts
ULPP702 and ULPP703.
References
[1] A. Sagara et al., Fusion Eng. Des. 83, 1690 (2008).
[2] A. Komori et al., J. Nucl. Mater. 241–243, 967 (1997).
[3] P. Grigull et al., Plasma Phys. Control. Fusion 43, A175
(2001).
Fig. 4. Top: CAD data for the modified divertor.
Bottom: CCD camera view of the modified divertor during a
discharge with H filter.
Fig. 5. Time evolutions of the stored energy and NB injection
power (top), line-averaged density and fueling amount
(middle), and neutral pressure in the modified and unmodified
divertors (bottom) during a density ramp-up discharge.
Stellarator News -4- August 2011
[4] A. Komori et al., Nucl. Fusion 45, 837 (2005).
[5] S. Masuzaki et al., J. Nucl. Mater. 363–365, 314 (2007).
[6] N. Ohyabu et al., Phys. Rev. Lett. 97, 055002 (2006).
[7] S. Masuzaki et al., Fusion Eng. Des. 85, 940 (2010).
[8] M. Shoji et al., Contrib. Plasma Phys. 48, 185 (2008).
[9] D. Reiter et al., Fusion Sci. Technol. 47, 172 (2005).
[10] M. Shoji et al., J. Nucl. Mater. 363–365, 827 (2007).
[11] Y. Kubota et al., Fusion Eng. Des. 75–79, 297 (2005).
[12] S. Masuzaki et al., Plasma Fusion Res. 6, 1202007
(2011).
Suguru Masuzaki
National Institute for Fusion Science
Toki, Japan
E-mail: masuzaki@LHD.nifs.ac.jp